Welcome and apologies
Members were welcomed to the meeting and provided introductions.
Apologies were noted.
All actions were complete.
CIS digital update
The board were presented with an update on the current status of the new digital format of the programme.
It was noted that study workers completed on 31st July with transition taking place over the last three weeks of July. All participants have been asked to take part in the digital format by the end of August.
The presentation provided the opt in rates for the ten cohorts approached at different times and it was noted that while uptake was good during the Private Beta phase, it improved with more recent cohorts.
The board were shown a table indicating the percentage of participants who opt in and then complete a questionnaire, with this data broken down by data of completion window starting and the day of completion window (1 to 14). This was broken down between Private and Public Beta. The data shows that the majority of participants complete their questionnaire early after being invited to do so.
Key IT delivery dates are on track to being met by their deadlines.
The key message for the mode effect analyses carried out so far is that there is little different between estimates of positivity from data collected during study worker home visits and remote data collection.
Multi-level Regression Post-stratification (MRP) models for positivity, comparing study worker data and digital collection data were presented. This analysis has been done for specific dates so does not match what has been published elsewhere by the study. When analysed at the level for England, there is no difference in trends.
Analyses broken down by region in England were presented. More variability would be expected here due to smaller sample size. The only clear difference is in the data from the North East of England where we can observe slightly different trends and there is more variability. There is overlap in the credible intervals between the study worker and the digital data. This is a small effect and nothing to raise concern.
Looking at a comparison of trend statements for English regions for a specific reference day. The one week trend shows very little differences between the two datasets. The two week trend shows some differences but this is not a concern. It was also noted that two weeks prior to the reference date, sample sizes would have been very small.
Analyses on the data from Devolved Administrations showed that there were no major differences for Wales and Scotland when comparing study worker and digital data. There is more variability for Devolved Administrations due to the smaller sample size. There was some difference in the trend, variability in the data from Northern Ireland. This may be because there was small study worker population in Northern Ireland.
When analysing positivity by age, it was noted that for school years 7-11 there is some effect in the point estimate. This is likely to due to the smaller sample size, further investigations are underway. So far, there does not appear to be big differences in other age groups, all credible intervals overlap for every other age group. Similar patterns can be observed when comparing contour plots for single year of age (England).
Next steps and upcoming publications were shared with the board. The first Management Information (MI) pack is being released tonight.
The Advisory Board noted these analyses are reassuring and show what a smooth transition to the digital format this has been, which is a great achievement. Credit was given both to the teams involved in this work and to the enthusiasm of the participants.
It was queried whether there was any indication of selective drop out or any particular cohort deciding not to continue into the digital format? At this point, no details have been collected on opt in rate but the team are looking into this and can feed back at a later date. It was noted however that there has been an increase in younger people participating in the survey since the move to digital.
The board noted that there had been an expectation with Census that older people would be less keen on digital contact but it was actually found that older people were happy to use digital forms of contact, perhaps as a result of experiences during lockdown. This survey seems to be seeing the same results. Younger people prefer the flexibility of the new approach but as the older cohort are digitally literate, we not seeing the drop out that may have occurred previously.
It was noted that there are benefits to having such an engaged cohort who have stayed with the study for so long.
Retention of participants was discussed. Attrition has been carefully monitored throughout the different stages of the transition. Additionally, there is an intensive comms programme in place to ensure we are making sure we respond to points raised by participants. We can also utilise experience across ONS from other online digital surveys.
Future Advisory Board Remit
Over the past 18 to 24 months, the focus has been on what the CIS can tell us about the pandemic. This Advisory Board has been critical in providing scrutiny and challenge and helping the think through thoughts and ideas. Now is the right time to grow the opportunity for this group and maximise the skills around the table.
It was acknowledged that it is important to consider how this board and CIS fits into the broader suite of evidence and support around population health surveillance.
The recently revised ONS Health Analysis Strategy was presented to board members. This strategy aims to delivery strategic ambitions and considers the drivers behind these. ONS acknowledge the importance of making the most of their rich data assets. There is capability within ONS to generate insights from these data assets. Work is ongoing to strengthen partnerships across ONS and wider government departments.
Two questions were posed to the board:
1. Should we re-visit the remit of this group? Should we cover a broader range of health beyond covid?
2. Does the membership reflect the remit?
Initial comments noted that focus of this board ties in with the survey itself and how this moves forward and develops.
ONS are working closely with the UK Health Security Agency (UKHSA), including joint publications and joint public health insights.
It was acknowledged that partnerships are important and should continue to be strengthened.
Board members agreed that a workshop would be useful to define potential future public health questions, and consider delivery costs and the sequencing of the decision making process. These discussions should also include other key surveys such as population based surveillance.
It was recognised that the future of health surveillance is likely to also consider mental health and wellbeing.
As there will likely be further developments on future plans by the time this group meets next (October), it was agreed that discussions around the remit of this group can be revisited then.
Any Other Business
The next meeting will take place on in October 2022, exact date to be confirmed.
- David Spiegelhalter (Chair)
- Maximillian Ayling
- Tali Boswell
- Tina Clapham
- Debbie Curtis
- Sir Ian Diamond
- Dan Dukes
- Jeremy Farrar
- Susan Hopkins
- Jonathon Laszlo
- Naomi Rogers
- Emma Rourke
- Ruth Studley
- Nick Taylor
- Thomas Waite
- Sian Hawkins
- Polly Marshall
- John Bell
- Peter Benton
- Jenny Harries
- Christopher Jewell
- John Newton
- Monika Preuss
- Steven Riley
- Sarah Walker